Electrooptical devices based on electrooptical materials such as lithium niobate (LiNbO3), KDP and KTP have been hereto described. See I. P. Kaminov et al., “Optical Fiber Telecommunications”, Vol. IIIB, Ed. by Academic Press (1997).
There are known electrooptical devices such as light modulators based on a D-shaped optical fiber and using a lithium niobate crystal. See W. Johnstone et al., “Fiber Optic Modulators Using Active Multimode Waveguide Overlays”, Electron. Lett., Vol. 27, No. 11, 894–896 (1991). D-shaped optical fiber, also called side-polished optical fiber, refers to an optical fiber having a D-shaped cross section. Light modulators of this type are usually produced on a plane-parallel plate of a quartz glass which is provided with a groove having a curvature typically from several dozens of centimeters to several meters. A single-mode or monomode optical fiber is glued into this groove. Then the plate side having the groove with the fiber glued therein is ground until this plane reaches the fiber core so that the fundamental mode (localized predominantly in the core) can penetrate through the reflective cladding to the polished surface. After this processing, the optical fiber section becomes D-shaped. See S. M. Tseng et al., “Side-Polished Fibers”, Appl. Optics, Vol. 31, No. 18, 3438–3447 (1992). The polished surface of the D-shaped optical fiber is coated with a thin transparent electrode layer of indium tin oxide (ITO) composition. Then a thin lithium niobate crystal is glued onto this electrode and ground to reduce the thickness to 20–30 microns. Finally, the second electrode is applied above the lithium niobate crystal layer.
The light modulator operates as follows. An external voltage applied to the electrooptical lithium niobate crystal changes the refractive index of the material and modifies the condition of resonance between the fundamental mode of the optical fiber and the guided modes of the lithium niobate layer. The resonance condition is essentially the condition of phase synchronism, or equal propagation constants of the guided modes of the planar optical waveguide with a lithium niobate core and the fundamental mode of the D-shaped optical fiber. When the modes are in resonance, the light signal is effectively pumped from the optical fiber into the lithium niobate crystal and the output signal intensity at the fiber end is decreased. If the applied voltage is changed so as to alter the refractive index of the lithium niobate crystal and break the resonance, the light passes through the D-shaped optical fiber without loss in intensity. In the prior art, a significant level of the output signal modulation is achieved by applying a voltage of 150 V to a 35 micron-thick control layer between ITO electrodes.
One disadvantage of the light modulator described above is that the manufacturing process for thin lithium niobate layers is very complicated. Further, the interelectrode distance determined by the thickness of the lithium niobate crystal is relatively large.
Optical switches using the same principle of operation have been described employing a layer of material with variable refractive index on the surface of a D-shaped optical fiber and a liquid crystal layer. See S. M. Tseng et al., “Low-Voltage Optical Fiber Switch”, Jpn. J. Appl. Optics, Part 2, Vol. 37, L42–L45 (1998). In the optical switches of this type, a voltage about 30 V is needed to break the resonance for an interelectrode distance of 13 microns. One disadvantage of this device is the relatively low operation speed determined by the slow response of the liquid crystal. The switching time is about 7 milliseconds and the liquid crystal cannot be reoriented by an ac voltage with a frequency of 100 Hz.
There are known electrooptical devices such as light modulators having charge carrier injectors. See E. R. Mustel et al., “Light Modulation and Scanning Methods”, Nauka, Moscow (1970). The light modulator of this type employs a layer of an electrooptical material representing an n-type semiconductor film on a substrate. The light propagates along this film which serves as the optical waveguide. Deposited above this n-type film is a layer of a p-type semiconductor, which forms a p-n junction. The device also contains a pair of electrodes, one in ohmic contact with the n-type semiconductor film and the other with the p-type semiconductor film, to which a control (dc or ac) voltage is applied. When a control voltage is applied to the p-n junction in the forward direction, the charge carriers (holes) are injected into the optical waveguide (n-type semiconductor film). The injection of holes into the optical waveguide increases the optical absorption of the material, thus modulating the light.
One disadvantage of this type of light modulators is the current-induced heating of the p-n junction, which requires taking special measures to thermally stabilize the entire device. Another disadvantage is the limitation imposed on the modulation frequency by the mechanism of light modulation employed in this device. Indeed, the lifetime of the minority carriers injected through the p-n junction is limited, usually to about 10−6 seconds for the holes. For this reason, the light modulators guided by the minority carrier injection can operate only at frequencies up to 105–106 Hz. The electric current passed through the optical waveguide must be of sufficiently large density. This requirement poses limitations on the system dimensions. The greater the size of the device, the higher the current required to maintain the density on a level necessary for the device operation. A further disadvantage related to the electric current passage is the large energy consumption, which increases with the current value.
There are known electrooptical devices which contain a layer of a material whose optical properties change depending on the applied electric field strength. See WO 00/45202. One example of such material is ferroelectric ceramics. Ceramic materials possessing ferroelectric properties usually exhibit the phenomenon of birefringence. Thus, the ceramic layer is an electrooptical material and the applied electric field can control the device. Owing to a combination of the ferroelectric and electrooptical properties of the material, this system can be employed for controlling and modulating light signals in fiber optic communication systems, nonlinear optical devices, and electrooptical devices such as modulators, shutters, and frequency multipliers, etc.
The observed optical effects are related to orientation or reorientation of the domain polarization vector in an applied electric field. As a result, the optical axes of the ceramic grains are oriented or reoriented as well. The reorientation of domains in the electrooptical ceramic material under the action of an applied electric field is accompanied by the development of mechanical stresses perpendicular to the field direction.
One disadvantage of the ferroelectric ceramics is that they retain orientation of the domain polarization vector for an arbitrarily long time after switching off the film. Therefore, additional measures have to be taken in order to restore the initial state, such as applying control pulses with opposite polarity and half amplitude, mechanically deforming the ceramic substrate, and applying a high-frequency electric field of small amplitude. This property of the ferroelectric ceramics complicates the control system of the electrooptical devices.
Another disadvantage of the ferroelectric ceramics is the difficulty of ensuring a fast operation speed. Indeed, an increase in the light modulation rate at a given modulation efficiency requires increasing the control voltage amplitude. This fact and the delayed electrooptical response in such materials are related to the energy consumption for the formation and reorientation of the domain walls. For example, at electric pulse duration of about 2 μs, the pulse amplitude must be two times greater than the quasistatic control voltage; to reduce the pulse duration to 1 μs, the pulse amplitude must be three times greater, and so on.
Additional disadvantage is the fatigue inherent in the ferroelectric ceramic materials. Straining a ceramic material in the range corresponding to the spatial modulation of light (e.g., at the expense of partial repolarization) encounters difficulties related to the deformation character of the field-induced polarization. For this reason, repeated on-off cycles of an electric field, especially of large strength (above 5 kV/cm), lead to the accumulation of a residual deformation. This residual deformation decreases the optical contrast of modulated light, which is manifested by irreversible polarization of the electrooptical ferroelectric ceramic layer.
Another disadvantage of the above device is extremely strong temperature dependence of the characteristics of a ferroelectric layer. Temperature variations lead to changes in the optical properties of the control device. In order to exclude the temperature drift, it is necessary to provide the control device with a thermal stabilization system, which increases the energy consumption, complicates the device, and increases the cost of production.
A significant disadvantage of the device employing ferroelectric ceramics is the probability of phase distortions introduced into the data processed as a result of strong deformation of the ceramic plate and the inverse piezoelectric effect. The presence of defects and internal stresses leads to degradation of the properties of such materials which are extremely sensitive to manufacturing process parameters, making production of the devices a difficult task.
There are known electrooptical devices based on organic materials. See U.S. Pat. No. 5,172,385 to Forrest et al. and L. M. Blinov, “Electro- and Magneto-optics of Liquid Crystals”, Nauka, Moscow (1978), pp. 115, 351, 352. The devices of this type contains two electrodes which are either both transparent if the system operates in the beam transmission mode, or transparent and reflecting, if the system operates in the beam reflection mode. An electrooptical material layer placed between the electrodes represents a liquid crystal, the thickness of which (interelectrode distance) is determined by sealing spacers. The electrodes are deposited onto glass substrates.
A large number of the chemical classes of organic molecules provides for a broad spectrum of materials which can be effectively used in fiber optics, integrated optics, and optical communications.
There are classes and groups of organic substances of various chemical natures, composed of the molecules or molecular chains such as phthalocyanines, polyacetylenes, aromatic hydrocarbons, conjugated polymeric systems, etc. that possess dielectric, semiconducting, and even metallic properties. A common feature of these molecules is the presence of superstructures. There are known organic films based on polymeric materials (U.S. Pat. Nos. 4,204,216; 4,663,001; 4,269,738; 5,104,580; 3,775,177; F.R. Patent No. 2,583,222), salts of linear polyaniline compounds (U.S. Pat. No. 4,025,704), phthalocyanine derivatives (U.S. Pat. Nos. 5,525,811; 6,051,702), organic dyes (U.S. Pat. No. 3,844,843), and porphyrins (U.S. Pat. Nos. 3,992,205; 3,935,031), which are widely used in modern electronic devices as the layers generating charge carriers in the course of photoelectron processes in photovoltaic devices (U.S. Pat. No. 4,164,431), solar cells (U.S. Pat. No. 3,844,843), and polarization devices (U.S. Pat. 5,172,385).
There are various known methods for the formation of organic films and creation of anisotropic film structures, for example Langmuir-Blodgett technique (U.S. Pat. No. 5,079,595), molecular beam epitaxy, etc. However, optical devices employing liquid-crystalline molecular compounds possess a number of disadvantages, in particular, require specially prepared substrates, alignment layers, or high-vacuum conditions for the obtaining of highly ordered and clean structures. Special and advanced technologies are employed and even these often cannot ensure the obtaining of films possessing a certain type of order and ensuring required optical anisotropy.
There are known electrooptical devices using an electrooptical material of a liquid crystal (host) matrix with dispersed organic dye (guest) molecules. See I. K. Vereshchagin et al., “Introduction to Optoelectronics”, p. 173, Vysshaya Shkola, Moscow (1991), L. M. Blinov et al., “Electrooptical Effects in Liquid Crystal Materials”, p. 182, Springer-Verlag, New York (1994). The control device operates on the guest-host interaction principle and is structurally analogous to that described above, comprising two electrodes which are either both transparent if the system operates in the beam transmission mode, or transparent and reflecting, if the system operates in the beam reflection mode. An electrooptical material layer placed between the electrodes represents a liquid crystal doped with dye molecules. The thickness of this layer (interelectrode distance) is determined by sealing spacers placed between the electrodes deposited onto glass substrates. The molecules of liquid crystal and dye are oriented in the same direction parallel to the alignment layers. In the absence of an applied voltage, a light polarized in the long axis of dye molecules is absorbed and no signal is transmitted through the optical device. This absorption in the dye is related to the fact that the electric field of the light polarized parallel to the long axis of the dye molecules drives electrons to oscillate between the ends of the molecule, thus consuming the beam energy. An external voltage applied to the electrodes creates an electric field in the liquid crystal. This field rotates the liquid crystal molecules and, hence, the dye molecules (due to the guest-host effect) so that the long axis of the dye molecules becomes perpendicular to the plane of polarization of the light beam. In this cease, electrons in the dye molecules are not forced to move by the electric field of the light beam. Therefore, the light is not absorbed in the liquid crystal layer and the beam is transmitted through the optical device without significant losses.
One disadvantage of the above optical device is the relatively low operation speed of, which is characterized by a switching time on the order of 0.1 second. This device operates poorly at reduced temperatures, under which conditions the operation speed further sharply reduces. The device has insufficient working life, which amounts to about 104 hours.